Despite the fact that hypomania being the hallmark of the disorder, patients spend far more time in and experience greater distress from their depressed states. Therefore, effective treatment for BP II depression is perhaps the most important issue in BP II.
Unfortunately, the majority of patients with BP II depression are inadequately treated. The National Comorbidity Survey Replication (9,282 subjects) has shown that only about 16% of patients with BP II received appropriate medication which was defined as given lithium/valproate, anticonvulsants or antipsychotics, while 60% received no medication.[5
] Similarly, data from the Jorvi Bipolar Study[39
] found that only 44% of subjects with BP II were treated with an anticonvulsant or lithium. Patients with BP II were significantly more likely to receive treatment with an antidepressant compared to subjects with BP I. Only 31% of patients with BP II were considered to be receiving adequate pharmacotherapy. In a community sample of newly diagnosed patients with BP II (n
=1 001), 55.5% were prescribed an antidepressant (65% of them had antidepressants as monotherapy), compared to 31% who were prescribed lithium, an anticonvulsant, or an antipsychotic.[40
The best available evidence for the acute treatment of bipolar depression comes from studies on quetiapine. There are now four large RCTs demonstrating the efficacy of quetiapine monotherapy in combined groups of patients with BP I or II depression: BOLDER I[41
] and II,[42
] and two additional eight-week RCTs, EMBOLDEN I[43
] and II.[44
] These four trials included substantial numbers of patients with BP II depression: BOLDER I (n
=181) and II (n
=170), EMBOLDEN I (n
=303) and II (n=262). In patients with BP II depression in BOLDER I and EMBOLDEN I, improvements in MADRS were numerically but not statistically significant at endpoint (week 8), although they were significant at various weekly visits. In contrast, the BOLDER II and EMBOLDEN II trials showed significant benefits in the patients with BP II. In addition, a post hoc pooled analysis of the patients with BP II depression from both BOLDER trials (n
=351) found that both doses of quetiapine demonstrated significant benefits as early as week 1, which were sustained throughout the 8 weeks.[45
] Two subanalyses of the BOLDER I data showed that among patients with BP II depression, quetiapine was effective in patients with rapid cycling,[46
] but anxiety scores (HAM-A) were not significantly improved.[47
] However, in the pooled analysis of BOLDER I and II, the changes in HAM-D, HAM-A, and CGI were significantly greater for both quetiapine groups versus placebo, and Quetiapine 600 mg/day was effective in both rapid and nonrapid cycling depression.[26
] Based on statistically significant improvements in two RCTs and numerically superior improvements in two additional trials, quetiapine monotherapy can now be recommended as a first-line treatment.
Other atypical antipsychotics have also been studied but no controlled studies are available so far with antipsychotics in BP II disorder. Olanzapine in combination with fluoxetine was found to be effective in BP I depression,[48
] but it is unclear whether it would work in BP II as well. On the other hand, an 8-week open trial of Ziprasidone monotherapy in 20 patients with BP II depression found the drug at a relatively low dose appears to be a rapid, effective, and generally well-tolerated treatment for BP II patients experiencing major depression. There was significant improvement in depression scores within 1-2 weeks, which were sustained to end of treatment.[49
] However, larger placebo-controlled trials are needed to confirm these findings.
Evidence from several open studies supports the effectiveness of lithium, lamotrigine, and valproate in the acute treatment of patients with BP II depression. A randomized open study comparing lithium and lamotrigine in BP II acutely depressed patients concluded that both were effective and there were no major treatment response differences between them.[50
] A 16-week, open, randomized trial assessed the efficacy of lithium (n
=56) or lamotrigine (n
=46) monotherapy in patients with acute BP II depression.[45
] Mean MADRS scores significantly decreased from baseline in both groups (Lamotrigine from 28.9 to 12.5 and lithium from 29.9 to 15.2), and there was no differences between the two treatments. There were no differences in response between patients with rapid cycling (72% of patients) or without, although there was a high dropout rate in the rapid cycling group (42% of patients).
Valproate is bot only used as monotherapy, but also as augmentation. A 12-week open trial of valproate sodium monotherapy in 19 BP II depressed outpatients showed a statistically significant response rate (P
<0.001) of 63%.[51
] It was also shown to be effective in a 7-week, open trial in 28 patients with BP II depression.[52
] Response was statistically similar with monotherapy (45%, n
=21) and adjunctive therapy (71%, n
=7). Although these data are suggestive, the sample size was too small and the results require confirmation from well-designed RCTs with larger number of subjects.
Much controversy surrounds the use of antidepressants in BP II depression.[9
] In BP II patients it is very important to control depressive episodes. The challenges for clinicians in managing patients with BP II is to treat acute episodes of depression without causing switches into hypomania. There may be situation that antidepressant treatment may be useful.[55
] Unfortunately the treatment of BP II disorder with antidepressants is still an understudied area. Most recommendations for the treatment of BP II depression are derived from findings of studies that have included both BP II and BP I patients. Like in the treatment with antidepressant (especially tricyclic antidepressant) in BP I patients, the risk of switching from depressive to hypomanic states when treated with antidepressants is also a concern in the use of antidepressant in BP II patients. However, the risk-benefit ratio for antidepressant use in BP II is still an unresolved issue. Whether the same risks and pitfalls associated with treating depression in BP I readily apply to BP II has not been conclusively demonstrated. Furthermore, the risk of hypomanic switch or cycle acceleration with antidepressants in patients with BP II is less than in those with BP I is still controversial. The risk of antidepressant-induced cycle acceleration has been reported by some authors to be more likely in BP II patients than their BP I counterparts.[53
] However in this study, the cases were taking the older generation (heterocyclic) antidepressants. In more recent studies in which patients were taking the newer generation antidepressants (eg, SSRI, venlafaxine, and buproprion), BP II patients have a lower switch rate compared to BP I patients.[56
There is some evidence for antidepressant monotherapy including fluoxetine,[58
], and citalopram.[62
] In a small, 9-month, randomized, crossover trial involving ten treatment-naive patients with BP II, patients received SSRI treatment had a significant reduction in depression severity, percentage of days depressed or high, and percentage of days impaired, without illness destabilization, when compared with placebo.[62
] Furthermore, a post hoc analysis of a placebo-controlled RCT of antidepressant monotherapy in 248 unipolar and 62 BP II patients found that both groups benefited comparably from active treatment, with no switch noted in the BP II patients.[63
Amsterdam and his colleagues have specifically addressed the use of antidepressants in BP II patients in three studies.[58
] The largest of these studies[58
] compared acute and continuation phase treatment with fluoxetine monotherapy in 89 BP II patients with age and gender matched, and 661 unmatched, unipolar (UP) patients. Although all subjects met criteria for a major depressive episode according to semistructured clinical interviews based on the structured clinical interview for DSM III-R, BP II patients were identified by retrospective chart review, presumably based on DSM-IV criteria. Subjects were part of a larger study, comprised of a 12-week, open label, treatment phase and a 50-week, double blind, placebo-substitution, relapse-prevention phase, designed to test the efficacy and safety of fluoxetine in the prevention of depressive relapse. Outcome measures were based on HAM-DI7 scores, with response defined as 50% reduction. Fluoxetine was effective in both short-term treatment of the depressive episode and in relapse prevention for patients in remission over a 1-year follow-up period. Short-term treatment was shown to be similarly efficacious for BP II and UP patients; a nonsignificant trend toward earlier reduction in HAM-D17 scores was observed in the BP II group. Likewise, survival analyses at 26, 50, and 62 weeks, demonstrated similar relapse rates between the two groups. It is worth highlighting that, by the 26th, 50th, and 62nd weeks the BP II patients numbered 28, 19, and 8, respectively (with comparable figures in the matched unipolar group). Manic switch episodes, ascertained retrospectively, were reported for 3.8% (three of 80) BP II patients, no matched UP patients and two of 661 unmatched UP patients during short-term treatment, and 3.6% (one of 28) BP II and 0.8% (two of 241) unmatched UP patients during the relapse-prevention phase. None of the manic switch episodes apparently met DSM criteria for mania. However, it should be noted that this retrospective evaluation may not have detected all cases of antidepressant-induced hypomania. Furthermore, the dropout rate was high. By the end of the relapse-prevention phase (62nd week) the sample size had dropped to only eight patients, a number far too small on which to base strong conclusions that UP and BP II responded similarly, both in terms of relapse rates and the induction of hypomania.
Venlafaxine has also been evaluated. In a small, 6-week, prospective trial,[60
] 30 UP and 16 BP II depressed patients were randomly assigned to receive once or twice daily venlafaxine monotherapy following a 1-week placebo lead in. A semistructured SCID interview was used to diagnose major depressive episodes; at entry, and after the first week, subjects had baseline HAM-D21 scores of 20. It is not clear how the diagnosis of BP II was established. Outcome measures were based on the HAM-D, Montgomery-Asberg depression rating scale (MADRS) and CGI, completed weekly for the first month and then again at 6 weeks. Although similar overall efficacy between the two groups was shown, BP II patients who completed the trial demonstrated a statistically significant more rapid reduction in their HAM-D (P
<0.03) and MADRS (P
<0.02) scores. However, when patients who did not complete the study (nine unipolar and two BP II) are accounted for by a last observation carried forward analysis; a nonsignificant trend is observed instead. No episodes of venlafaxine-induced switching were observed. Another 12-week, open, randomized trial in 83 patients with BP II depression found that there were higher response and remission rates with venlafaxine compared to lithium.[64
] Discontinuation rates were significantly lower with venlafaxine compared to lithium, and there was no evidence of hypomanic switch in either group.
On the other hand, bupropion does not have positive result as in the case of fluoxetine and vanlafaxine. In a 16-week RCT of adjunctive Bupropion in 20 patients with BP II depression who had an inadequate response to 8 weeks of Lamotrigine found no differences between bupropion and placebo on either depression or mania scores.[65
] However, the number of subjects was too small to make any definite conclusion on the effectiveness of the drug.
All these studies have shown that there is a low rate of treatment emerge hypomanic switch with antidepressant monotherapy in BP II patients. Actually, the overall lower propensity to develop hypomanic symptoms of BP II to I patients has also been shown.[66
] According to a meta-analysis of available data for antidepressant treatment in BP II patients, the rate of treatment emerge hypomanic switch during acute treatment may be intermediate between BP I and unipolar depression.[67
The STEP-BP study,[68
] comparing adjunctive antidepressants (bupropion or paroxetine) plus lithium or valproate and lithium or valproate alone, for up to 26 weeks, included 114 patients with BP II. In the combined sample (BP I and BP II) as well as between BP I and BP II patients, rates of durable recovery (8 consecutive weeks of euthymia) were comparable for adjunctive antidepressants and lithium or valproate alone. Although the antidepressants did not increase the risk of manic switch, it did not have any added benefit when the patient is already receiving a mood stabilizer. Hence, antidepressant treatment may not be useful for every patient and perhaps should be reserved for those patients who have inadequate response to other mood stabilizers such as lithium, anticonvulsant, and newer generation antipsychotic medication.
As far as the switch into mania with antidepressant use is concerned, all the larger studies suggest that this risk is quite modest, at least when combined with a mood-stabilizing medication, and seem to be generally lower in BP II than in BP I patients. However, data on the long-term effect of use of antidepressant in BP II is scarce. Therefore, the appropriate role for antidepressants in the acute and long-term treatment of BP II disorder will need to be defined by future well-controlled studies.